The implementation of pneumococcal conjugate vaccines (PCVs) into the routine childhood immunization schedule in US and then in Europe has markedly reduced the burden of invasive pneumococcal disease (IPD), including meningitis.1 In France, 7-valent PCV (PCV7) was recommended in 2003 for children younger than 2 years considered at high risk of pneumococcal infection, whether intrinsic (underlying conditions) or extrinsic (children in a daycare center, with a sibling or breastfed for <2 months), and in 2006 for all children at age 2, 3 and 4 months with a booster at 18 months (3 + 1 schedule). In 2008, the schedule was changed for all children at age 2 and 4 months with a booster at 12 months (2 + 1 schedule).2
In the PCV7 era, the frequency of vaccine serotype (VT) meningitis significantly decreased.3 However, this reduction has been eroded to varying degrees by an increase in frequency of meningitis because of non-VTs, including serotype 19A.3 In 2010, French authorities recommended routine infant immunization with 13-valent PCV (PCV13) to replace PCV7, without a catch-up for older children except those at high risk of IPD (2 + 1 schedule).4 The increase in pneumococcal disease because of 19A, 7F and 3 (3 of the 6 additional pneumococcal serotypes in PCV13 compared with PCV7) was halted or reversed by the introduction of the PCV13.1,5 Nevertheless, as for other vaccines, cases of vaccine failure (VF) or vaccine breakthrough (VBT) have been reported.6,7 Although comorbidity is a risk factor, these cases occurred mostly in healthy children.8 Only a few studies have focused on VBT and VF since the PCV13 implementation.
Thirteen years ago, the Pediatric Infectious Diseases Group of the French Pediatric Society established an active surveillance network to monitor the clinical and biological features of pneumococcal meningitis (PM).3,5 Here, we analyze cases of VBT and VF of PM VT since 2003, after PCV implementation in France.
MATERIALS AND METHODS
From 2003 to 2013, 233 pediatrics wards working with 168 microbiology departments throughout France were asked to report all cases of bacterial meningitis. The methodology of our surveillance network was previously described.5 The following diagnostic criteria were used for Streptococcus pneumoniae meningitis in children aged 1 day to 15 years: clinical signs associated with positive cerebrospinal fluid (CSF) culture and/or positive CSF antigen testing, positive CSF polymerase chain reaction findings and/or positive culture of a normally sterile body site associated with CSF pleocytosis (>10 cells/μL). Vaccine status was documented in the immunization chart of each child and checked in the investigator’s medical dossier for each patient. S. pneumoniae infection was identified by standard methods in the microbiology laboratory of each hospital. Isolates were serotyped at the National Reference Center for Pneumococci by agglutination with latex particles sensitized with antisera from the Statens Serum Institute (Copenhagen, Denmark). To characterize pneumococcal VBT and VF meningitis, we used the definitions of Park et al7: VBT defined as a PM in a child who had received ≥1 PCV dose and for which the pneumococcal isolate was a PCV serotype, and VF defined as a subset of VBT occurring in a child who had completed the series of PCV or who had received the appropriate catch-up doses. Children whose PCV vaccinations were up-to-date for age but who had not completed a full age-appropriate series before their PM were considered VBT but not VF. Moreover, a 2-week interval was required between PCV administration and disease onset to be considered VBT or VF.7 Meningitis in children who received PCV7 doses and at least 1 PCV13 as a booster dose was also considered PCV13 VF because 1 dose of PCV13 in a child older than 12 months induced similar immunity for the 6 additional serotypes as 3 doses of PCV13 in infants.9 Data were entered with use of 4D v6.4 and analyzed by use of Stata SE 9.1 (Stata Corp., College Station, TX) and Statview II (Abacus concept).
From 2003 to 2013, 1233 cases of PM in children were diagnosed: 943 children (76.5%) from 2003 to 2010 in the PCV7 era and 290 (23.5%) from 2011 to 2013 in the PCV13 era. VBT accounted for 39 (3.2%) cases: 24 (2.0%) with at least 1 PCV7 dose and 15 (1.2%) with at least 1 PCV13 dose. All cases but one were in children younger than 5 years at diagnosis [mean age, 19.5 months (range, 2.6–150.1 months); 41% girls]. The underlying conditions were reported for 11 patients: congenital meningeal breach (n = 2), posttraumatic meningeal breach (n = 2), cochlear implant (n = 1), recurrent meningitis without immunodeficiency (n = 1) and immunodeficiency (n = 5). Among the 5 patients with immunodeficiency, we found acute leukemia in the consolidation phase of treatment (n = 1), X-linked agammaglobulinemia (n = 1) and hypogammaglobulinemia (IgG, n = 2; IgG2 subclass, n = 1).
In the PCV7 era, 6 of 943 children (0.6%) were considered to have VF (Table 1). One child died of PM because of serotype 14, despite PCV7 vaccination and absence of immunodeficiency. In the PCV13 era, 2 of 290 children (0.7%) were considered to have VF (Table 1). Table 2 shows the 31 cases of VBT but not VF. In the PCV7 and PCV13 eras, 13 of 18 and 9 of 13 children, respectively, were adequately vaccinated for age. Serotype 19F was the most frequent cause of the VF, even after the introduction of PCV13. VF and VBT were more frequent after the beginning of vaccination with PCV7 and PCV13, and its frequency decreased thereafter (see Fig., Supplemental Digital Content 1, http://links.lww.com/INF/C196).
This study is the largest focusing on VBT and VF PM after PCV7 and PCV13 implementation. In our series, serotype 19F was most frequently responsible for PCV7 failure. In the Ladhani et al10 study, serotype 6B was responsible for 34% of VF PM cases and serotype 19F for 30.2%. PCV7 is known to induce antibodies with lower avidity for serotype 19F than other PCV7 serotypes, which could explain the greater number of VFs with this serotype.11 Furthermore, serotype 19F is the only PCV7 VT not completely eradicated from the nasopharyngeal microbiome several years after PCV7 implementation.12 The antibody titers produced are lower for serotype 6B than other PCV7 VTs, and only 58.4–88% of children achieved a protective threshold (vs. >95% for other PCV7 VTs) after the primary series. Then, after early first injections, the mean antibody titers increase moderately to become acceptable after the toddler dose.11 This kinetics pattern may explain why serotype 6B VF precociously occurs after the first series and always before the toddler dose. Similarly, a lower vaccine response to serotype 23F has been reported, with the protective threshold reached for antibody titers against this serotype (like serotype 6B) only after the booster dose. Serotype 14 was otherwise considered highly immunogenic.11
Serotype 19A was the main PCV13 VT involved in PM in PCV13 recipients. This result is supported by Kaplan et al,1 who reported the only 2 other cases of PM with PCV13 VF. The antibody titers against serotype 19A induced by PCV13 and their avidity estimated by opsonophagocytic assay (OPA) titers reached an acceptable level in several studies of immunogenicity. However, Bryant et al11 showed that the avidity of antibodies was lower for serotype 19A than many other VTs. As OPA titers did not predict protection, the VF due to serotype 19A appears difficult to explain.13
Underlying conditions that likely altered host defenses against S. pneumoniae infection were rarely found in our cases of PCV VF, and Kaplan et al1 reported that VF occurred mainly in children younger than 6 months or with comorbidity.7
The largest number of our VF cases occurred within 2 years after the implementation of PCV, before a decrease in number (even if the period of observation is lower after PCV13). These trends could be explained by herd immunity in the pediatric population after the decrease in PCV VT carriage.12 Indeed, as for Haemophilus influenzae serotype b vaccination, herd immunity plays an important role in conjugate vaccine effectiveness.
Our study has several limitations. First, although our bacterial meningitis network has been established for 12 years and the number of pediatric and bacteriologic departments participating and the response rates of these services have remained stable throughout the period considered, some cases of VF might have escaped our surveillance. Second, we did not systematically investigate the immunological status (including antibody values by enzyme-linked immunosorbent assay and OPA for each VT) of all children with VF. Indeed, Gaschignard et al14 recently showed that children with IPD should undergo immunological investigations because primary immunodeficiencies may be discovered in up to 26% of cases (children >2 years old). However, our study was a large nationwide survey of 1233 cases of PM in children, providing data not reported by any other surveillance system.
In conclusion, PM because of VT despite PCV vaccinations are rare and mostly occur in children younger than 2 years without an underlying condition or more rarely in older children with an underlying condition. Three years after PCV13 implementation, serotype 19A PM is not completely eradicated. Assessing the immune response after vaccination in children with PCV failure would help better understand the underlying mechanisms.
We thank all pediatricians and microbiologists of the “Observatoire National des Méningites” who participated in this study.
1. Kaplan SL, Barson WJ, Lin PL, et al. Early trends for invasive pneumococcal infections in children
after the introduction of the 13-valent pneumococcal conjugate vaccine
. Pediatr Infect Dis J. 2013;32:203–207
2. Institut de veille sanitaire. Vaccination Schedule for 2008. Recommendations from the Haut Conseil de la Santé Publique. 2008;16–17:129–47 Available at: http://www.invs.sante.fr/beh/2008/16_17
. Accessed April 21, 2008
3. Hsu HE, Shutt KA, Moore MR, et al. Effect of pneumococcal conjugate vaccine
on pneumococcal meningitis
. N Engl J Med. 2009;360:244–256
4. Institut de veille sanitaire. French Vaccinal Calendar Recommendations. 2010 Available at: http://www.invs.sante.fr/beh/2010/14_15
. Accessed April 22, 2010
5. Levy C, Varon E, Picard C, et al. Trends of pneumococcal meningitis
after introduction of the 13-valent pneumococcal conjugate vaccine
in France. Pediatr Infect Dis J. 2014;33:1216–1221
6. Heininger U, Bachtiar NS, Bahri P, et al. The concept of vaccination failure. Vaccine. 2012;30:1265–1268
7. Park SY, Van Beneden CA, Pilishvili T, et al.Active Bacterial Core surveillance team. Invasive pneumococcal infections among vaccinated children
in the United States. J Pediatr. 2010;156:478–483.e2
8. Harboe ZB, Valentiner-Branth P, Ingels H, et al.Danish Pneumococcal Surveillance Collaborating Group. Pediatric invasive pneumococcal disease caused by vaccine serotypes following the introduction of conjugate vaccination in Denmark. PLoS One. 2013;8:e51460
9. . Licensure of a 13-valent pneumococcal conjugate vaccine
(PCV13) and recommendations for use among children
—Advisory Committee on Immunization Practices (ACIP), 2010 MMWR Morb Mortal Wkly Rep. 2010;59:258–261
10. Ladhani SN, Slack MP, Andrews NJ, et al. Invasive pneumococcal disease after routine pneumococcal conjugate vaccination in children
, England and Wales. Emerg Infect Dis. 2013;19:61–68
11. Bryant KA, Block SL, Baker SA, et al.PCV13 Infant Study Group. Safety and immunogenicity of a 13-valent pneumococcal conjugate vaccine
. Pediatrics. 2010;125:866–875
12. Cohen R, Levy C, Bonnet E, et al. Dynamic of pneumococcal nasopharyngeal carriage in children
with acute otitis media following PCV7 introduction in France. Vaccine. 2010;28:6114–6121
13. Andrews NJ, Waight PA, Burbidge P, et al. Serotype-specific effectiveness and correlates of protection for the 13-valent pneumococcal conjugate vaccine
: a postlicensure indirect cohort study. Lancet Infect Dis. 2014;14:839–846
14. Gaschignard J, Levy C, Chrabieh M, et al. Invasive pneumococcal disease in children
can reveal a primary immunodeficiency. Clin Infect Dis. 2014;59:244–251